determining factors of cross-country dispersion in life satisfaction

Determining factors of cross-country dispersion in life satisfaction:

evidence from Europe

(Work in progress)

Daphne Nicolitsas

To be presented in Session 4.2 - Parents-Offspring relations and life satisfaction∗

June 19, 2016

Abstract

Life satisfaction scores appear pronouncedly negatively skewed in some countries - in general countries

with high average life satisfaction - but much more uniformly distributed in other countries. The deter-

minants of these cross-country differences in dispersion is the issue this paper investigates. Following the

work of Hamermesh (2001) the hypothesis being tested is that dispersion is lower in countries in which the

probability of a discrepancy between the outcomes of individuals’ lives and their expectations is narrow.

A number of descriptive characteristics appear to be consistent with this hypothesis. A more formal test

of this hypothesis is done by investigating the association between the difference of actual remuneration

from the remuneration predicted by using observable individual characteristics (fitted income) with life

satisfaction. The results so far suggest that the hypothesis put forward cannot be rejected

JEL classification: I39, J17, J28, J31, M52

Keywords: life satisfaction; fairness; expectations

1 Introduction

Across country differences in average life satisfaction are a familiar feature of cross-country studies on subjective

well-being. The satisfied Danes and the not so satisfied citizens of Eastern European countries are by now

almost cliches. Differences in per capita income are part of the explanation although the low average subjective

well-being measure for France, lower than in the Czech Republic, cannot be attributed to this.1

∗University of Crete; [email protected]. Support through grant KA4446 by the University of Crete is gratefully acknowledged.

Participants at the Annual Meeting of the Society of Scottish Economists in April 2016 and at the Research Seminar Series at the

Economics Department of Freie University in May 2016 provided useful comments on a slightly earlier version.1See, for example, Figure 2.2. in the World Happiness Report 2016, Vol. I.

1

Countries, however, differ in other aspects of life satisfaction besides the average score. Life satisfaction

scores are pronouncedly negatively skewed in certain countries — in general countries with high average life

satisfaction — but much more uniformly distributed in other countries. Explanations for the cross-country

differences in the dispersion of life satisfaction is the issue this paper investigates. Starting from the premise

that individuals assess their satisfaction from life by comparing outcomes with expectations (Kahneman, 1999)

the hypothesis put forward here is that in countries in which the distribution of life satisfaction scores is

negatively skewed, individuals are less dissatisfied with outcomes. This is so because their expectations are

more frequently met but also because, in the absence of the perception of generalized unfair treatment, the

probability that they will be treated fairly is higher. The framework is based on the work of Hamermesh (2001)

and is presented in Section 2.

The empirical analysis is conducted using the European Social Survey (ESS) database described in

Section 3. Section 4 presents some descriptive information while Section 5 presents and discusses a more formal

analysis. Finally, Section 6 summarizes and concludes.

2 Framework

Hamermesh (2001) aims to show the existence or otherwise of a link from the inequality of earnings to the

dispersion of job satisfaction. He argues that job satisfaction depends on workers’ expectations about earnings

and working conditions. Depending on when expectations are formed and how swiftly these are revised,

Hamermesh distinguishes between four hypotheses which he subsequently tests. The results suggest that

workers’ job satisfaction is especially responsive to surprises in the returns to observable skills but less so to

surprises in the returns to unobservables.

In a separate strand of literature, Schwarze (2008) argues that individuals’ life satisfaction depends on

ex ante income uncertainty.

It is the above two strands of the literature that we aim to combine here.

One way of modelling the determination of life satisfaction is to assume that the difference from average

life satisfaction depends on the size of the discrepancy between expectations and realizations. Individuals for

which realizations are worse than expectations are likely to be disappointed and exhibit lower than average life

satisfaction than those for which realizations either match or exceed expectations [the asymmetry needs to be

explained].

In a schematic way for individual i we have:

Si − S = f(Ri − Ei) (1)

S satisfaction, S is average life satisfaction, R realized (actual) outcome, E expected outcome. Re-

alizations and expectations concern a number of dimensions: activity status, earnings, health, social activity,

family life. where f > 0 if Ri < Ei and f = 0 if Ri > Ei.

2

A substantive is the form of function f : a linear function would imply that satisfaction decreases at

the same rate independently of the size of the discrepancy between realizations and expectations. If instead

we assume that f is a quasi-concave function (f′′> 0) then we are assuming that satisfaction decreases at an

increasing rate as the discrepancy between realizations and expectations increases. We further assume that

the sensitivity of the reaction of satisfaction to the discrepancy depends on the extent to which the society is

deemed to be fair - in effect this is a measure of ex ante income uncertainty.

The way expectations are formed is also important - as a simplification, however, we will assume here

that they are formed based by the observable skills (education, tenure) that one possesses. The role of family

background will be investigated.

3 The European Social Survey

The European Social Survey (ESS) is a cross-national survey of individuals aged 15 and over resident within

private households. The survey is conducted every two years since 2002, and by 2014 had completed 7 rounds.

The survey, however, is not longitudinal. While the sample of countries participating in the survey has not

been constant over time, a sample of 16 countries is present in all 7 rounds.2 The survey collects information

on attitudes, beliefs and behavioural patterns using the same questionnaire in each country.3

The questions to elicit the subjective well-being (SWB) measures are presented in Table 1. The survey

contains data on two types of measures, following the classification introduced by Deaton and Stone (2013),

evaluative and hedonic measures. Evaluative measures are based on individuals’ rankings of life satisfaction and

happiness on a 0 to 10 scale with 0 being ‘extremely dissatisfied (unhappy)’ and 10 being ‘extremely satisfied

(happy)’. Measures of job satisfaction and of satisfaction from the work-life balance achieved are also reported

in Rounds 3-6. Hedonic measures, which have been used by inter alia, Kahneman et al. (2006), ask individuals

about the length of time they have felt calm and relaxed, active and cheerful during the last one or two weeks.

According to Deaton and Stone (2013) the two types of measure are associated with different types of

income: evaluative measures are more closely related to permanent income while hedonic measures are more

closely related to transitory measures of income. As the issue being investigated here is structural rather

than conjunctural, we will be using the evaluative measures. [As am extemsion we will, however, also look at

associations using the hedonic measures as we do not have measures of permanent income.]

The survey also contains measures of income. In all rounds (non-equivalised household) income coded

in intervals is available. However, this is usable only from Round 4 (2008) onwards: only from then do

intervals correspond to country income deciles. Prior to 2005 intervals had not been adjusted for cross-country

differences. In two rounds (Round 2 and Round 5), however, individual remuneration income (gross4 of taxes

and social security contributions) in absolute figures is also recorded. [Missing values do exist and the lack of

2Table 12 provides information on the composition of the sample by round.3Information on methodological issues (e.g. sampling, sample sizes by country, dealing with non-response etc) can be found at

the following link http://www.europeansocialsurvey.org/methodology/index.html.4In Round 2 data on net pay has also been collected.

3

Table 1: The survey questions underlying the reported SWB measures

Round Variable Question

Evaluative measures

All rounds Life satisfaction All things considered, how satisfied are you with your

life as a whole nowadays? Please answer using this card,

where 0 means extremely dissatisfied and 10 means ex-

tremely satisfied

Round 3 Life satisfaction How satisfied are you with how your life has turned out so

far? Please use this card were 0 is labelled as ‘Extremely

dissatisfied’ and 10 as ’Extremely satisfied’

Round 3 Standard of living And how satisfied are you with your present standard of

living? Please use this card were 0 is labelled as ‘Ex-

tremely dissatisfied’ and 10 as ’Extremely satisfied’

All rounds Happiness Taking all things together, how happy would you say you

are? Please use this card were 0 is labelled as ‘Extremely

unhappy’ and 10 as ’Extremely happy’

Rounds 3,5,6 Job satisfaction All things considered, how satisfied are you with your

present job? Please use this card were 0 is labelled as

‘Extremely unsatisfied’ and 10 as ’Extremely satisfied’

Rounds 3,5,6 Work-life balance

satisfaction

How satisfied are you with the balance between the time

you spend on your paid work and the time you spend on

other aspects of your life? Please use this card were 0 is

labelled as ‘Extremely unsatisfied’ and 10 as ’Extremely

satisfied’

Hedonic measures

Rounds 3,6 Calmness Time in the last week felt calm and peaceful? Range of

1-4 with 1 labelled ‘None or almost none of the time’ and

4 labelled ‘All or almost all of the time’

Rounds 2,5 Calmness I have felt calm and relaxed in the last two weeks. Range

of 1-6 with 1 labelled ‘All of the time’ and 6 labelled ‘At

no time’

Rounds 2,5 Cheerful I have felt cheerful and in good spirits in the last two

weeks. Range of 1-6 with 1 labelled ‘All of the time’ and

6 labelled ‘At no time’

Rounds 2,5 Active I have felt active and vigorous in the last two weeks.

Range of 1-6 with 1 labelled ‘All of the time’ and 6 la-

belled ‘At no time’

Source: European Social Survey (ESS).

4

Table 2: ESS measures of whether justice is being served

Variable Values Round

Courts’ decisions are unduly influenced by political pressure Agreement 1-5 5

Courts protect rich & powerful over ordinary people Agreement 1-5 5

Frequency of impartial & fair court decisions Scale 0-10 5

Assessment of quality of courts’ job Scale 1-5 5

Courts treat eveyone the same Scale 0-10 6

Importance for democracy that courts treat all equally Scale 0-10 6

Source: European Social Survey (ESS).

randomness in replies remains an issue.] Information is also available on the period covered by this pay (hourly,

daily, monthly or annual income). Table 13 in the Appendix shows the number of observations, in Round 5,

for which information on absolute remuneration levels is available in each country together with the number of

all employed individuals. Respondents to the ESS survey are also asked whether they think the remuneration

paid reflects the effort they put in the job. Finally, the ESS also contains information on whether individuals

assess their household income as being adequate.

As the focus on this paper is on the comparison between actual and expected outcomes and the feeling

of injustice, the ESS database appears suitable since it contains information about individuals’ perceptions of

courts effectiveness. These measures could proxy individuals’ views on fairness. Table 2 presents details on

these variables.

It is possible, as argued by Schwarze (2008), that well-being is influenced negatively by ex ante income

uncertainty. Ex ante income uncertainty is likely to be higher when fair treatment is not prevalent. A separate

measure of uncertainty is also included in the ESS database: individuals are asked to report whether they feel

their current job is secure.

4 Descriptive information on the distribution of life satisfaction

measures

Differences in average life satisfaction are illustrated in Figure 1. These are by now well-known and the rankings

of countries on this measure appear to remain relatively constant over time [reference].

What is less known is that substantial cross-country differences exist in the distribution of life satisfac-

tion within each country. Figure 2 shows the distribution of life satisfaction in each country. The data refer to

the 15 countries for which ESS data for 2014 are available. The countries have been organized in three groups

according to the cumulative percentage of the population with a life satisfaction score of 7 or below. Countries

with the lowest share of individuals with a score below 7 are in the top most row, countries with the highest

share of individuals with a score below 7 are in the lowest row of the Figure. A feature that emerges from this

5

Figure 1: Average life satisfaction across countries (2014)

Figure is that in countries in the top row of Figure 2, which are countries with high average life satisfaction

scores, the distribution of life satisfaction is very negatively skewed.

The first issue we address when trying to investigate the above cross-country differences is whether

this is a conjunctural phenomenon. It appears that this is not the case; Figure 5 in the Appendix shows that

cross-country differences in the distribution of life satisfaction according to the 2004 ESS data is very similar

to the picture for 2014.

The second issue has to do with composition effects. More specifically, we seek to establish whether

cross-country differences in the distribution of life satisfaction reflect differences in the tightness of the labour

market or differences in per capita income. In order to investigate whether the increased concentration of

individuals at the higher end of the life satisfaction distribution in, for example, Norway is due to the very low

unemployment rate there we look at the distribution of life satisfaction only for employed individuals. Figure

3 shows the data. The picture that emerges is similar to that of Figure 2 which presented the distribution for

the population as a whole.

As life satisfaction and household income are positively related (see, inter alia, Deaton, 2011) a possible

reason behind Figure 2 is that countries with a negatively skewed distribution of life satisfaction are, in general,

higher income countries. Differences in per capita income could not, however, explain why, for example, the

distribution of life satisfaction in Germany is less negatively skewed than in Finland nor why the distribution

of life satisfaction in France or Ireland is less negatively skewed than in Poland.5 While comparisons of the life

5Data on per capita income per country are presented in Table 14 in the Appendix. Furthermore, Section A in the Appendix

shows the income deciles in each country. These are the thresholds used for the income intervals in the ESS questionnaire for

Round 7 (2014).

6

Figure 2: Distribution of life satisfaction scores (2014)

Figure 3: Distribution of life satisfaction scores (2014)- Employed individuals only

7

satisfaction of individuals with the same level of income across countries is not possible, Table 3 shows average

life satisfaction by decile in each country. The figures clearly suggest that average life satisfaction increases

with income and that the cross-country dispersion in average life satisfaction is lowest for the highest income

deciles. However, the difference in average life satisfaction between the highest and the lowest income deciles

is highest for countries with lower average life satisfaction. Table 4 presents the percentage of individuals with

a score below or equal to 7 in each income decile in each country. However, in contrast to the average life

satisfaction measure here the coefficient of variation appears to be highest for the high income groups.

Table 3: Average life satisfaction by income decile, 2014

Country 1 2 3 4 5 6 7 8 9 10 Average (10)-(1)

AT 6.6 7.2 7.0 7.2 7.3 7.7 8.0 7.6 7.5 8.3 7.3 1.7

BE 6.7 7.1 7.2 7.2 7.2 7.6 7.6 7.7 7.7 8.0 7.4 1.3

CH 7.4 7.1 7.9 8.1 8.3 8.4 8.2 8.3 8.5 8.4 8.1 1.0

CZ 5.6 6.1 6.0 6.3 6.8 6.9 7.0 6.8 6.8 7.5 6.5 1.9

DE 6.2 6.6 6.9 7.1 7.4 7.6 7.7 7.9 8.0 8.2 7.3 2.0

DK 8.0 7.9 8.0 8.1 8.3 8.3 8.5 8.5 8.7 8.8 8.8 0.8

FI 7.2 7.4 7.6 7.9 7.9 8.1 8.1 8.1 8.4 8.4 8.4 1.2

FR 5.3 5.3 6.0 6.4 6.5 6.5 7.0 7.2 7.3 7.7 7.5 2.4

IE 6.7 6.6 6.9 7.2 7.1 7.4 7.2 7.4 7.5 8.1 7.9 1.4

NL 6.7 7.0 7.2 7.3 7.6 7.7 7.8 8.0 8.1 8.1 8.1 1.4

NO 7.4 7.6 7.9 8.0 8.0 7.8 8.2 8.1 8.3 8.2 8.2 0.8

PL 6.1 6.6 6.7 6.6 6.8 7.2 7.1 7.5 7.3 7.8 7.9 1.7

SE 7.3 7.1 7.6 7.8 7.7 7.9 7.8 8.1 8.3 8.4 8.2 1.1

SI 5.3 5.7 6.1 6.4 6.9 7.1 6.9 7.5 7.6 7.5 7.7 2.2

Std. dev. 0.85 0.75 0.68 0.64 0.56 0.52 0.52 0.46 0.55 0.37 0.58 0.50

Coef.Variation 0.13 0.11 0.096 0.089 0.076 0.069 0.068 0.060 0.070 0.046 0.073 0.33

Source: ESS, 2014

The cross-country dispersion in life satisfaction appears to be less pronounced for youth. Table 5 shows

the percentage of individuals with a score below or equal to 7 by age group. It is clear that countries’ differences

with respect to the distribution of life satisfaction scores exists for each age group. However, differences are

less pronounced for younger age groups as witnessed by the fact that the coefficient of variation of the % of

individuals with a score or equal to 7 is lower for these groups. This appears to be consistent with the view

that young individuals have not been disillusioned. Admittedly, this difference across age groups could also

arise for other reasons: earnings of youth are probably more compressed both within and between countries.

We turn next to see whether we can find associations between life satisfaction and first, the extent to

8

which individuals perceive the society they live in as fair and second, the extent to which remuneration received

differs from remuneration expected.

Table 4: % of individuals with a life satisfaction of 7 or below by income decile, 2014

1 2 3 4 5 6 7 8 9 10

Denmark (DK) 24.6 26.8 25.8 24.3 22.2 19.3 15.3 15.0 8.4 8.7

Switzerland (CH) 37.0 46.7 30.8 27.2 24.2 19.1 23.8 16.8 17.1 20.7

Norway (NO) 43.9 40.5 33.3 28.5 29.2 29.6 19.8 24.6 20.4 18.4

Finland (FI) 41.6 37.3 42.8 28.9 27.2 21.9 19.8 23.7 12.5 12.6

Netherlands (NL) 59.0 52.8 53.7 51.5 42.2 38.6 29.0 25.9 22.6 17.2

Sweden (SE) 46.2 44.9 42.2 33.6 36.4 35.1 33.8 29.2 20.7 19.9

Germany (DE) 63.3 56.6 53.7 46.6 41.5 40.2 38.7 36.0 26.1 17.4

Belgium (BE) 57.7 49.2 50.0 50.3 52.6 38.7 37.9 32.7 31.1 24.8

Austria (AT) 59.2 50.2 50.3 49.1 45.4 39.5 29.3 30.5 40.0 22.2

Poland (PL) 64.4 59.3 58.9 56.3 55.1 44.6 47.5 44.4 48.9 28.3

Ireland (IE) 61.2 58.6 57.1 54.1 54.0 42.5 47.1 42.7 39.0 33.3

Slovenia (SI) 77.9 74.2 68.0 66.4 55.3 52.1 50.5 40.6 37.3 46.7

France (FR) 72.9 70.2 70.7 65.5 59.4 60.7 54.7 50.0 39.6 38.7

Czech Rep. (CZ) 74.6 64.5 65.1 69.0 61.7 56.7 53.2 57.6 56.0 36.7

St. dev. 15.4 12.9 13.8 15.5 13.6 12.9 13.4 12.4 14.0 10.8

Coef.Variation 0.27 0.25 0.27 0.33 0.31 0.34 0.37 0.37 0.47 0.44

Source: ESS, 2014

5 Analysis

5.1 Fairness in general

We start from the issue of fairness. In order to test the hypothesis that life satisfaction is higher when individuals

feel they are or will be treated more fairly, we use the information in the ESS database which pertains to the

perception of fairness. As indicated in Section 3 the ESS contains data on how individuals view the reward

of justice in the country they are residing, whether they are being fairly treated and appropriately paid.

Unfortunately, these variables are not available in every round. Most of these variables are available for Round

5 (see Table 2) and this is the Round we use in the estimates below. Figure 4 presents the distribution for

the variable that shows disagreement with the view that the courts protect the rich and powerful. Individuals’

views are coded on a 1 to 5 scale, with 1 corresponding to Strong agreement and 5 with Strong disagreement.

9

Table 5: % of individuals with a life satisfaction of 7 or below by age group, 2014

Age groups

15-19 20-29 30-39 40-49 50-59 60-64 65-74 75+

Denmark (DK) 10.7 21.5 20.4 19.8 19.0 15.4 14.5 17.2

Switzerland (CH) 22.8 30.7 28.2 29.4 26.8 19.0 22.8 23.8

Norway (NO) 29.4 35.3 31.1 31.1 27.6 27.5 21.2 32.6

Finland (FI) 26.8 33.3 25.2 29.5 25.9 22.7 26.2 25.1

Netherlands (NL) 37.7 37.8 34.6 36.8 42.2 38.8 35.5 44.1

Sweden (SE) 30.7 37.9 34.5 31.3 30.4 25.8 26.2 32.5

Germany (DE) 36.5 39.8 41.2 39.4 45.7 42.2 37.1 35.4

Belgium (BE) 27.6 41.8 41.5 45.6 40.9 40.3 43.2 39.7

Austria (AT) 17.5 37.3 45.1 49.5 45.5 42.9 46.6 49.1

Poland (PL) 47.9 43.2 42.1 51.0 59.8 63.9 54.3 54.5

Ireland (IE) 39.3 58.4 53.5 55.8 58.3 53.3 48.8 47.1

Slovenia (SI) 37.9 54.5 49.1 56.4 70.6 62.3 60.2 66.7

France (FR) 37.1 50.9 60.5 61.7 63.7 61.1 61.8 60.7

Czech Rep.(CZ) 47.9 48.7 59.9 59.9 69.1 66.1 61.7 67.7

Estonia (EE) 42.1 55.6 50.8 64.9 74.8 66.9 74.3 71.4

St. dev. 10.6 10.2 12.2 14.1 18.4 18.2 17.9 17.1

Coef. Variation 32.3 24.5 29.7 32.0 39.4 42.2 42.4 38.4

Source: ESS, 2014

10

Figure 4: Disagreement with the view that courts protect rich & powerful

In order to simplify the analysis somewhat I have aggregated the life satisfaction variable into 4

categories from the original 11 categories; the first category includes scores 0 to 5, the second category includes

score 6 and 7, the third category corresponds to score 8 and the last two scores to category 4. The main features

of the picture do not change significantly although some of the detail is lost; Table 6 shows the distribution of

the grouped variable by country.

Table 7 shows the marginal effects from estimating the determinants of life satisfaction. The focus

is here on two variables: the assessment on whether the pay received is appropriate and the view on the

impartiality of the courts. The estimated equations also include the following variables that typically appear

in life satisfaction equations: age, age squared, gender, marital status, years of education, health condition,

social activity. The coefficients on these variables are as expected: age is U shaped (although age is not always

significant), men are less satisfied, married individuals have a higher life satisfaction and so do individuals with

higher social activity than their peers and finally education does not appear to make much of a difference in

satisfaction once all other variables have been included. The equations also include information on the decile of

household income the individual belongs to, the feeling about the adequacy of household income as well as an

indication of the ease of borrowing money. These variables are in most cases significant and with the expected

signs.

The appropriateness of pay variable ranges from 1 to 5. A value of one (five) indicates that the

individual strongly agrees (strongly disagrees) with the view that conditional on effort and achievement the

11

Table 6: Distribution of life satisfaction on an aggregated scale, 20100

Country 1 (0-5) 2 (6-7) 3 (8) 4 (9-10)

DK 5.7 14.2 27.8 52.4

CH 7.7 17.5 29.7 45.1

NO 9.1 20.2 31.7 39.0

FI 8.0 18.0 35.8 38.3

NL 7.7 25.8 41.7 24.8

SE 10.4 18.6 32.5 38.4

DE 23.1 24.1 27.0 25.8

BE 11.3 28.8 35.0 24.9

PL 24.4 24.7 25.6 25.3

IE 32.9 28.6 21.4 17.1

CZ 33.7 30.6 23.0 12.7

FR 36.1 28.6 20.2 15.0

SI 27.3 22.7 25.6 24.3

EE 32.6 28.0 21.4 17.9

Source: ESS, 2010.

pay received is appropriate. The variable has been used in the regression as 5 separate dummy variables (with

one dummy - the one corresponding to plain agreement - used as the reference group) each corresponding to

a different degree of agreement. The impartiality of courts variable, as already mentioned above, also ranges

from 1 to 5. However, I use this variable as a continuous variable in the estimated equations.

The marginal effects (with standard errors in brackets) reported in Table 7 show the probability of

being included in the first and last category respectively of the aggregated life satisfaction variable if the

independent variable takes the value 1 (for the variable showing the different levels of appropriateness of pay)

and for a one unit increase in the value of the impartiality of courts variable. Marginal effects in bold are

significant at either the 1, 5 or 10% level.

The impact of one unit change in the courts’ variable seems to substantially increase the probability of

being in the high category in most countries; Finland, Greece and Slovenia are exceptions in that this variable

is not significant. Furthermore, the impact appears to be greater in countries in which the % of individuals

in the highest category appears to be the largest while the largest impact regarding the lowest categories are

found in France and Poland.

Turning to the appropriateness of pay variable it appears to be the case that dissatisfaction with pay

is an important component of life satisfaction in countries with low level of life satisfaction.

12

Table 7: Marginal effects fom an ordered probit of aggregated life satisfaction

Country Courts Given effort and achievements, pay is appropriate

Agree Neither agree Disagree Disagree

strongly nor disagree strongly

DK-1 -0.0067 (0.0031) -0.0143 (0.0114) 0.0160 (0.0090) 0.0012 (0.0075) -0.0116 (0.0119)

DK-4 0.0399 (0.0176) 0.0857 (0.0666) -0.0958 (0.0499) -0.0069 (0.0445) 0.0691 (0.0699)

CH-1 -0.0090 (0.0053) -0.0273 (0.0185) 0.0103 (0.0172) 0.0196 (0.0145) 0.0266 (0.0365)

CH-4 0.0266 (0.0156) 0.0810 (0.0540) -0.0304 (0.0505) -0.0583 (0.0421) -0.0789 (0.1080)

NO-1 -0.0101 (0.0052) -0.0406 (0.0213) 0.0225 (0.0116) -0.0022 (0.0110) 0.0150 (0.0232)

NO-4 0.0316 (0.0159) 0.1267 (0.0655) -0.0703 (0.0365) 0.0068 (0.0342) -0.0470 (0.0723)

FI-1 -0.0037 (0.0033) -0.0294 (0.0146) 0.0321 (0.0101) 0.0284 (0.0087) -0.0061 (0.0164)

FI-4 0.0159 (0.0140) 0.1256 (0.0591) -0.1369 (0.0384) -0.1211 (0.0333) 0.0259 (0.0701)

NL-1 -0.0107 (0.0042) -0.0100 (0.0127) -0.0038 (0.0088) -0.0019 (0.0078) -0.0255 (0.0212)

NL-4 0.0445 (0.0155) 0.0419 (0.0526) 0.0158 (0.0367) 0.0080 (0.0324) 0.1066 (0.0855)

SE-1 -0.0145 (0.0060) 0.0298 (0.0239) 0.0223 (0.0139) 0.0296 (0.0136) 0.0436 (0.0246)

SE-4 0.0398 (0.0158) -0.0819 (0.0656) -0.0613 (0.0379) -0.0814 (0.0364) -0.1198 (0.0660)

DE-1 -0.0149 (0.0082) 0.0119 (0.0385) 0.0509 (0.0209) 0.0289 (0.0209) 0.0972 (0.0286)

DE-4 0.0176 (0.0097) -0.0140 (0.0454) -0.0601 (0.0246) -0.0341 (0.0245) -0.1147 (0.0337)

UK-1 -0.0208 (0.0087) -0.0422 (0.0361) 0.0466 (0.0277) 0.0658 (0.0214) 0.0971 (0.0422)

UK-4 0.0254 (0.0105) 0.0516 (0.0442) -0.0570 (0.0337) -0.0805 (0.0258) -0.1187 (0.0516)

IE-1 -0.0354 (0.0144) -0.0774 (0.0514) 0.1765 (0.0381) 0.0868 (0.0451) 0.0661 (0.0771)

IE-4 0.0246 (0.0098) 0.0537 (0.0355) -0.1224 (0.0283) -0.0602 (0.0311) -0.0458 (0.0536)

CZ-1 -0.0415 (0.0137) -0.0539 (0.0651) -0.0361 (0.0330) -0.0259 (0.0378) 0.0343 (0.0538)

CZ-4 0.0224 (0.0076) 0.0292 (0.0355) 0.0195 (0.0180) 0.0140 (0.0206) -0.0186 (0.0292)

FR-1 -0.0239 (0.0132) -0.0437 (0.0539) 0.1191 (0.0368) 0.0724 (0.0326) 0.1042 (0.0526)

FR-4 0.0143 (0.0079) 0.0261 (0.0324) -0.0713 (0.0225) -0.0433 (0.0195) -0.0624 (0.0315)

GR-1 0.0107 (0.0181) 0.0899 (0.0977) 0.0574 (0.0465) 0.1214 (0.0497) 0.1592 (0.0783)

GR-4 -0.0037 (0.0064) -0.0315 (0.0341) -0.0201 (0.0161) -0.0426 (0.0183) -0.0558 (0.0276)

PL-1 -0.0381 (0.0125) -0.1595 (0.1020) 0.0657 (0.0335) 0.1262 (0.0291) 0.1199 (0.0463)

PL-4 0.0405 (0.0132) 0.1696 (0.1081) -0.0698 (0.0355) -0.1342 (0.0300) -0.1275 (0.0486)

EE-1 -0.0302 (0.0121) -0.0307 (0.0604) 0.0803 (0.0311) 0.0698 (0.0295) 0.0711 (0.0567)

EE-4 0.0229 (0.0092) 0.0232 (0.0457) -0.0607 (0.0237) -0.0528 (0.0226) -0.0538 (0.0429)

SI-1 0.0162 (0.0169) -0.0720 (0.0910) -0.0306 (0.0411) 0.1269 (0.0389) 0.0546 (0.0798)

SI-4 -0.0162 (0.0169) 0.0719 (0.0903) 0.0306 (0.0411) -0.1267 (0.0395) -0.0545 (0.0796)

The equations also include the variables: age, age2, health status, social activity intensity,

years of education, household income deciles, adequacy of household income, ability to borrow.

Effects in bold suggest significance at the 1%, 5% or 10% level. Standard errors in parentheses.

13

5.2 Discrepancy between actual and expected remuneration

We next turn to the other measure of the discrepancy between expectations and outcomes. As, already

mentioned, we first proxy the expected pay with the fitted variable of a typical earnings equation. Table 8

presents estimates of the earnings equations estimated. The estimates are based on data for 2010 (Round 5)

of the ESS survey for all countries except Norway and the UK for which the regression performance on the

basis of data for 2010 was very poor and the data for 2004 (for both the wage and the following life satisfaction

regression) have been used instead. The fitted values from these regressions are used to proxy the wages

expected by each individual.

Table 8: Wage regressions

Dependent variable: log of gross earnings

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15)

Variables CH NO FI NL SE DE BE PL IE UK SI FR CZ EE GR

Age 0.044** 0.083*** 0.039*** 0.043*** 0.046*** 0.141*** 0.036* 0.054*** 0.013 -0.015 0.050* 0.028* 0.028* 0.006 0.012

(0.016) (0.013) (0.010) (0.012) (0.007) (0.021) (0.015) (0.015) (0.034) (0.044) (0.021) (0.012) (0.011) (0.014) (0.020)

Age2 -0.000** -

0.001***

-

0.000***

-

0.000**

-

0.000***

-

0.002***

-0.000 -

0.001***

-0.000 0.000 -

0.001**

-0.000* -

0.000**

-0.000 -0.000

(0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.001) (0.000) (0.000) (0.000) (0.000) (0.000)

Men 0.426*** 0.244*** 0.179*** 0.304*** 0.141*** 0.361*** 0.267*** 0.290*** 0.270* 0.343* 0.260*** 0.192*** 0.203*** 0.325*** 0.093

(0.065) (0.046) (0.035) (0.047) (0.027) (0.066) (0.050) (0.054) (0.128) (0.143) (0.058) (0.042) (0.034) (0.064) (0.061)

Establishment size - 1 to 10 emp. reference group

10-24 0.080 -0.051 0.058 0.110 0.033 0.118 0.123 0.102 0.233 -0.246 0.139 0.086 -0.015 0.217** -0.042

(0.079) (0.062) (0.043) (0.068) (0.040) (0.123) (0.076) (0.085) (0.164) (0.238) (0.108) (0.056) (0.044) (0.071) (0.078)

25-99 -0.016 0.052 0.135** 0.167** 0.094** 0.366** 0.090 0.077 0.367* 0.019 0.229** 0.121* 0.012 0.231** 0.020

(0.084) (0.060) (0.044) (0.063) (0.035) (0.116) (0.069) (0.071) (0.169) (0.221) (0.083) (0.052) (0.045) (0.071) (0.089)

100-499 0.192 0.067 0.108** 0.241*** 0.115** 0.477*** 0.104* 0.102* 0.485** -0.017 0.061 0.190*** 0.082 0.414*** 0.091

(0.105) (0.066) (0.053) (0.068) (0.041) (0.114) (0.071) (0.074) (0.172) (0.222) (0.085) (0.057) (0.051) (0.089) (0.141)

500+ 0.299* 0.066 0.254*** 0.237** 0.243*** 0.359** 0.166* 0.328*** 0.260 -0.169 0.090 0.106* 0.094 0.435* 0.295*

(0.134) (0.072) (0.061) (0.076) (0.045) (0.114) (0.080) (0.092) (0.218) (0.227) (0.090) (0.063) (0.064) (0.169) (0.147)

Type of organization - private sector co. reference group

Government 0.071 -0.147 0.042 -

0.133***

0.136 -0.074 -0.018 0.176 0.041 0.080 -0.059 -0.045 -0.098

(0.102) (0.077) (0.093) (0.038) (0.114) (0.086) (0.113) (0.250) (0.110) (0.065) (0.060) (0.091) (0.100)

Continued on Next Page. . .

Table 8: Wage regressions (continued)



Other pub-

lic sector

-0.200 -

0.179***

0.004 -

0.168**

0.193 -0.194* -0.034 -0.117 0.027 0.052 -0.057 -0.071 0.073

(0.125) (0.051) (0.077) (0.052) (0.099) (0.082) (0.114) (0.184) (0.094) (0.076) (0.064) (0.121) (0.121)

SOE 0.020 -0.124 -0.042 -0.012 0.171 -0.097 -0.092 -0.351 -0.011 0.108* -0.144* -0.097 -0.008

(0.135) (0.063) (0.103) (0.107) (0.148) (0.092) (0.100) (0.317) (0.074) (0.072) (0.064) (0.117) (0.118)

Self-

employed

0.065 -0.115* -

0.268**

-0.048 0.181 -0.137 0.293** 0.271 -0.022 -0.101 0.196** -0.277 -0.033

(0.126) (0.058) (0.083) (0.055) (0.109) (0.105) (0.111) (0.580) (0.138) (0.111) (0.069) (0.162) (0.084)

Other 0.003 -0.022 0.038 -0.052 -0.242 -0.194 -0.297* 0.204 -0.048 -0.398* -0.141 -0.231 -0.068

(0.228) (0.079) (0.111) (0.078) (0.240) (0.143) (0.131) (0.359) (0.259) (0.176) (0.096) (0.160) (0.238)

Marital status - married reference group

Not mar-

ried

0.022 0.066 -0.027 -0.021 -0.014 0.053 0.034 0.047 -0.015 0.057 -0.091 -0.012 -0.052 0.077 -0.024

(0.064) (0.041) (0.032) (0.044) (0.025) (0.074) (0.051) (0.052) (0.126) (0.131) (0.056) (0.035) (0.031) (0.055) (0.057)

No of super-

visees

0.007* 0.001 0.003** 0.003** 0.002* -0.000 0.001 0.000 0.004 0.003 0.005** 0.005** 0.009** 0.001 -0.000

(0.003) (0.000) (0.001) (0.001) (0.001) (0.000) (0.001) (0.000) (0.005) (0.002) (0.002) (0.002) (0.003) (0.002) (0.000)

Experiernce 0.008* 0.005* 0.003 0.000 0.010* 0.006* 0.011*** 0.012 0.009* 0.011*** 0.007*** 0.002 0.014***

(0.003) (0.002) (0.003) (0.001) (0.005) (0.003) (0.003) (0.007) (0.003) (0.002) (0.002) (0.003) (0.004)

Constant 6.704*** 8.676*** 6.689*** 6.088*** 6.433*** 3.265*** 6.519*** 4.730*** 9.254*** 10.242*** 5.763*** 6.323*** 5.819*** 5.730*** 6.242***

(0.353) (0.568) (0.204) (0.273) (0.175) (0.500) (0.341) (0.338) (0.728) (1.424) (0.458) (0.251) (0.243) (0.335) (0.453)

Obs. 396 688 630 424 683 313 412 471 234 308 310 623 528 528 447

Continued on Next Page. . .

Table 8: Wage regressions (continued)



Adj. R2 0.501 0.405 0.520 0.511 0.543 0.554 0.482 0.461 0.370 0.321 0.569 0.476 0.405 0.390 0.271

Standard errors in parentheses, *** p<0.001, ** p<0.01, * p<0.05, p<0.10

In line with extended Mincerian earnings functions in the literature the estimates of the wage regressions

in Table 8 include variables capturing the degree of education (highest level of education completed) and to

tenure (number of years of working experience). The estimated equation captures the age-earnings profile using

the age and age squared variables. In addition given the differences in earnings across sectors and occupations

the regressions include sectoral and occupational dummies and indicators of the extent of supervisory roles.

As in some countries, marriage allowance increases earnings a dummy for marital status has been included.

Data have not been pooled as coefficients differ significantly across countries [formal test] although

results are as anticipated.

Using the fitted values from these regressions we calculate the discrepancy between the actual value

and the fitted wage. A positive value would suggest that the wage is higher than that expected on the basis of

observable characteristics and vice versa. [Care has to be taken given that conjunctural factors, not least due

to the crisis, might impact on the results for that year.]

The discrepancy thus calculated is then introduced as an explanatory variable in ordered probit regres-

sions on life satisfaction. Tables 9-11 show the coefficient estimates and marginal effects from these equations

(in fact this is a single table that refers to different countries, note the UK is missing for the time being). The

marginal effects refer to the relative probability of being in the first or the last category of the aggregated

measure of life satisfaction. What comes out from the estimates is that the discrepancy between the wage and

the fitted wage, in other words the returns to unobservable characteristics, appears to be significant in those

countries in which the low levels of life satisfaction have a higher concentration (Germany, Poland, Slovenia,

France, Estonia). The sign suggests that when the outcome is higher than the expected value the probability

of being at the lower end of the aggregated life satisfaction measure is depressed while the likelihood of being

at the high end of the aggregated life satisfaction measure increases. On the other hand, when the outcome is

worse than the expected value (i.e. wage discrepancy is negative) then this increases the probability of being

in the lowest level of the aggregated life satisfaction measure and decreases the probability of being in the

highest level of the aggregated life satisfaction measure. [But the ‘theory’ assumption of asymmetry is not

being tested!]

Table 9: Aggregated life satisfaction: Coefficient estimates & marginal effects from ordered probit estimates

Variable Denmark Norway Finland Netherlands

Coef. ME (1) ME (4) Coef. ME (1) ME (4) Coef. ME (1) ME (4) Coef. ME (1) ME (4)

Men 0.225 -0.0192 0.0814 0.0183 -0.0026 0.00625 -0.159 0.0142 -0.0560 -0.147 0.0111 -0.0431

Age -0.0227 0.00193 -0.0082 -0.0501 0.00709 -0.0171 0.0322 -0.00287 0.0113 -0.00788 0.000598 -0.00232

Age2 0.000371 0 0.000134 0.000646 0 0.00022 -0.0003 0 -0.00011 0.000148 0

Number of years of education (9-12 years reference group)

≤ 9 years 0.158 -0.0135 0.0571 0.466 -0.0661 0.159 0.141 -0.0125 0.0494 -0.126 0.00954 -0.0371

12 < y < 16 -0.0558 0.00476 -0.0202 -0.0627 0.00888 -0.0214 -0.0456 0.00407 -0.0161 -0.0168 0.00127 -0.00495

y > 16 0.125 -0.0107 0.0453 -0.141 0.0200 -0.0482 0.0398 -0.00355 0.0140 -0.0453 0.00344 -0.0134

Health status (Fair health is the reference group)

V. good 0.549 -0.0468 0.198 0.388 -0.0549 0.132 0.410 -0.0366 0.144 0.111 -0.00845 0.0329

Good -0.261 0.0223 -0.0944 -0.314 0.0445 -0.107 -0.543 0.0484 -0.191 -0.950 0.0721 -0.280

Bad -0.574 0.0490 -0.207 -0.295 0.0418 -0.101 -1.677 0.149 -0.591 -0.0555 0.00421 -0.0164

V. bad -0.962 0.0820 -0.348 -6.0376 0.855 -2.0574 -1.353 0.121 -0.477

Extent of social activity compared to peers (same as peers is the reference group)

Much less -0.465 0.0397 -0.168 -0.866 0.123 -0.295 -0.308 0.0274 -0.108 -0.456 0.0346 -0.134

Less -0.174 0.0149 -0.0630 -0.211 0.0299 -0.0718 -0.270 0.0241 -0.0953 -0.234 0.0178 -0.0691

More 0.0122 -0.00104 0.00442 0.219 -0.0311 0.0747 -0.156 0.0140 -0.0552 0.0567 -0.0043 0.0167

Much more 0.838 -0.0715 0.303 0.621 -0.0880 0.211 -0.0319 0.00285 -0.0112 0.172 -0.0130 0.0507

Wage diff. -0.129 0.0111 -0.0469 -0.00818 0.00116 -0.00279 0.199 -0.0177 0.0700 0.167 -0.0127 0.0493

The marginal effects concern the probability of being in the first or fourth category respectively.)

Coefficients in bold indicate significance at 1, 5 or 10% level.)


Variable Sweden Germany Poland Ireland


Men 0.00414 -0.00047 0.00143 -0.0764 0.0152 -0.0238 -0.290 0.0763 -0.0822 0.110 -0.0314 0.0164

Age -0.0203 0.00232 -0.007 -0.0514 0.0102 -0.0160 -0.0814 0.0214 -0.0231 -0.118 0.0337 -0.0177

Age2 0.000298 0 0.000102 0.000633 -0.00013 0.000197 0.000995 -0.00026 0.000283 0.00152 -0.00044 0.000228


≤ 9 years -0.103 0.0117 -0.0354 0.353 -0.0704 0.110 -0.829 0.218 -0.235 -0.23 0.0671 -0.0352

12 < y < 16 -0.168 0.0192 -0.0580 0.975 -0.194 0.303 0.0152 -0.004 0.00431 0.687 -0.196 0.103

> 16y -0.0675 0.00771 -0.0233 0.928 -0.185 0.289 0.223 -0.0588 0.0634 1.209 -0.345 0.181


V. good 0.632 -0.0722 0.218 0.442 -0.0881 0.138 0.390 -0.103 0.111 0.608 -0.174 0.0910

Good -0.383 0.0437 -0.132 -0.416 0.0828 -0.129 -0.292 0.0768 -0.0828 -0.140 0.0400 -0.0209

Bad -0.22 0.0251 -0.0758 -0.660 0.131 -0.205 -0.28 0.0745 -0.0803 0.149 -0.0425 0.0223

V. bad 0.0701 -0.008 0.0242 -1.129 0.225 -0.351 -5.740 1.512 -1.630 -4.557 1.301 -0.682


Much less -0.509 0.0581 -0.175 -0.168 0.0335 -0.0524 -0.111 0.0293 -0.0316 -0.287 0.0821 -0.04301

Less -0.266 0.0304 -0.0917 -0.0874 0.0174 -0.0272 0.0567 -0.0149 0.0161 -0.101 0.0290 -0.0152

More -0.0276 0.00315 -0.00951 0.0779 -0.0155 0.0242 0.134 -0.0353 0.0380 0.520 -0.148 0.0778

Much more 0.340 -0.0388 0.117 0.110 -0.0220 0.0343 0.00831 -0.00219 0.00236 1.908 -0.545 0.286

Wage discrepancy 0.110 -0.0125 0.0378 0.417 -0.0831 0.130 0.502 -0.132 0.143 -0.127 0.0362 -0.0191

The marginal effects concern the probability of being in the first or fourth category respectively.

Coefficients in bold indicate significance at 1, 5 or 10% level.


Variable Slovenia France Czech Republic Estonia


Men 0.0843 -0.0252 0.0224 0.116 -0.0396 0.0240 -0.236 0.0732 -0.0442 0.106 -0.0320 0.0228

Age 0.0382 -0.0114 0.0101 -0.0662 0.0226 -0.0137 -0.0060 0.0019 -0.0011 -0.0195 0.0059 -0.0042

Age2 -0.0005 0.0001 -0.0001 0.0008 -0.0003 0.0002 0.0001 0.0000 0.0000 0.0002 -0.0001 0.0000


≤ 9 years -0.171 0.0511 -0.0454 -0.0890 0.0303 -0.0184 -0.338 0.105 -0.0633 0.122 -0.0369 0.0263

12 < y < 16 0.359 -0.107 0.0953 0.211 -0.0719 0.0436 0.2735 -0.0847 0.0511 0.258 -0.0781 0.0556

> 16y 0.348 -0.104 0.0922 0.176 -0.0601 0.0364 0.344 -0.106 0.0642 0.4202 -0.127 0.0906


V. good 0.378 -0.113 0.100 0.416 -0.142 0.0861 0.447 -0.138 0.0835 0.731 -0.221 0.158

Good -0.314 0.0937 -0.0833 -0.124 0.0424 -0.0257 -0.599 0.186 -0.1129 -0.461 0.1406 -0.0994

Bad 0.304 -0.0907 0.0806 -0.982 0.335 -0.203 -0.412 0.128 -0.0770 -1.478 0.447 -0.31

V. bad 0.335 -0.100 0.0888 -5.0252 1.713 -1.038 -5.163 1.600 -0.965 (omitted) (omitted) (omitted)


Much less -0.0878 0.0262 -0.0233 -0.404 0.138 -0.0835 -0.560 0.174 -0.105 -0.256 0.0774 -0.0551

Less -0.0802 0.0239 -0.0213 -0.237 0.0809 -0.0490 0.0596 -0.0185 0.0111 -0.202 0.0613 -0.0437

More 0.0437 -0.0131 0.0116 0.0342 -0.0117 0.0071 0.567 -0.176 0.106 -0.204 0.0616 -0.0439

Much more 0.0168 -0.0050 0.0045 0.403 -0.137 0.0834 0.837 -0.2592 0.156 -0.155 0.0470 -0.0335

Wage discrepancy 0.565 -0.169 0.150 0.236 -0.0804 0.0488 -0.0568 0.0176 -0.0106 0.191 -0.0578 0.0412

The marginal effects concern the probability of being in the first or fourth category respectively.

Coefficients in bold indicate significance at 1, 5 or 10% level.

6 Summary & Conclusions

Cross-country differences in the distribution of life satisfaction are such that in some countries only very few

individuals have a score below the median on the life satisfaction rating ladder. In other countries, however,

the life satisfaction scores appear much more evenly distributed across the ladder. A potential explanation for

these cross-country differences is that in countries in which individuals expect to be treated fairly they are not

disappointed by outcomes. The reverse is true in countries in which justice is not always being served. The

results suggest that it is in countries in which there is a perception of lack of fairness that individuals’ life

satisfaction is especially responsive to deviations between actual and expected remuneration.

7 References

Deaton, A. (2011), ‘The financial crisis and the well-being of Americans’, Oxford Economic Papers, 64:1,

pp 1-26.

Deaton, A. and Stone, A.A. (2013) , ‘Two happiness puzzles’, American Economic Review: Papers &

Proceedings, 103(3): 591597.

ESS Rounds 1-7 : Norwegian Social Science Data Services, Norway Data Archive and distributor of ESS

data for ESS ERIC.

Hamermesh, D. (2001), ‘The changing distribution of job satisfaction’, Journal of Human Resources, 36(1):1-

30.

Helliwell, J., Layard, R., & Sachs, J. (2016), World Happiness Report 2016, Update (Vol. I). New York:

Sustainable Development Solutions Network.

Kahneman, D. (1986), ‘Objective happiness’ in Kahneman, D., E. Diener, N. Schwarz (eds.) Well-being:

the foundations of hedonic psychology. New York: Russell Sage Foundation.

Kahneman, D., A.B. Krueger, D. Schkade, N. Schwarz, A.A. Stone (2006), ‘Would you be happier

if you were richer?’, CEPS Working Paper No. 125.

Schwarze, J. (2008), ‘Subjective measures of economic well-being and the influence of income uncertainty’,

IZA DP. No. 3720.

UN (2016) , World Happiness Report, Vol. I.

Figure 5: Distribution of life satisfaction scores (2014)

Appendices

Appendix A Details on the ESS database

Appendix B Distribution of life satisfaction in other years

Appendix C Income per capita

Figure 6: Income deciles in each country (2014)

Table 12: EU Countries participating in each survey round and sample size

Country Round1 Round 2 Round 3 Round 4 Round 5 Round 6 Round 7

Austria 2,257 2,256 2,405 1,795

Belgium 1,899 1,778 1,798 1,760 1,704 1,869 1,769

Bulgaria 1,400 2,230 2,434 2,460

Switzerland 2,040 2,141 1,804 1,819 1,506 1,493 1,532

Cyprus 995 1,215 1,083 1,116

Czech Republic 1,360 3,026 2,018 2,386 2,009 2,148

Germany 2,919 2,870 2,916 2,751 3,031 2,958 3,045

Denmark 1,506 1,487 1,505 1,610 1,576 1,650 1,502

Estonia 1,989 1,517 1,661 1,793 2,380 2,051

Spain 1,729 1,663 1,876 2,576 1,885 1,889

Finland 2,000 2,022 1,896 2,195 1,878 2,197 2,087

France 1,503 1,806 1,986 2,073 1,728 1,968 1,917

UK 2,052 1,897 2,394 2,352 2,422 2,286

Greece 2,566 2,406 2,072 2,715

Hungary 1,685 1,498 1,518 1,544 1,561 2,014

Ireland 2,046 2,286 1,800 1,764 2,576 2,628 2,390

Italy 1,207 960

Lithuania 1,677 2,109

Luxembourg 1,552 1,635

Latvia 1,980

Netherlands 2,364 1,881 1,889 1,778 1,829 1,845 1,919

Norway 2,036 1,760 1,750 1,549 1,548 1,624 1,436

Poland 2,110 1,716 1,721 1,619 1,751 1,898 1,615

Portugal 1,511 2,052 2,222 2,367 2,150 2,151

Romania 2,146

Sweden 1,999 1,948 1,927 1,830 1,497 1,847 1,791

Slovenia 1,519 1,442 1,476 1,286 1,403 1,257 1,225

Slovakia 1,512 1,766 1,810 1,856 1,847

Table 13: Number of individuals with gross pay information & number of employed in Round 5

Country Pay informa-

tion available

Employed

indivisuals

Belgium 537 867

Bulgaria 562 915

Switzerland 590 920

Cyprus 311 553

Czech Republic 655 1,215

Germany 1,094 1,594

Denmark 758 876

Estonia 574 879

Spain 602 924

Finland 774 910

France 764 882

UK 882 1,199

Greece 558 1,061

Hungary 522 755

Ireland 853 961

Lithuania 398 616

Netherlands 614 1,026

Norway 892 984

Poland 557 887

Portugal 288 820

Sweden 786 853

Slovenia 361 655

Slovakia 462 834

Table 14: Per capita income in 2014, PPP

Country Income

Norway 67,100

Switzerland 59,160

Netherlands 48,860

Germany 47,460

Austria 47,380

Sweden 46,870

Denmark 46,850

Belgium 44,090

Ireland 42,830

Finland 40,630

France 40,100

United Kingdom 39,500

Slovenia 30,360

Czech Rep. 28,740

Estonia 27,490

Greece 27,050

Source: World Bank Databank.

determining factors of cross-country dispersion in life satisfaction

Documents